Swapped goal-conditioned offline reinforcement learning

1 code implementation • 17 Feb 2023 • Wenyan Yang, Huiling Wang, Dingding Cai, Joni Pajarinen, Joni-Kristen Kämäräinen

Offline goal-conditioned reinforcement learning (GCRL) can be challenging due to overfitting to the given dataset.

Offline RL reinforcement-learning +1

Prioritized offline Goal-swapping Experience Replay

no code implementations • 15 Feb 2023 • Wenyan Yang, Joni Pajarinen, Dinging Cai, Joni Kämäräinen

In goal-conditioned offline reinforcement learning, an agent learns from previously collected data to go to an arbitrary goal.

reinforcement-learning Reinforcement Learning (RL)

Monolithic vs. hybrid controller for multi-objective Sim-to-Real learning

1 code implementation • 17 Aug 2021 • Atakan Dag, Alexandre Angleraud, Wenyan Yang, Nataliya Strokina, Roel S. Pieters, Minna Lanz, Joni-Kristian Kamarainen

Our findings show that the training of a hybrid controller is easier and obtains a better success-failure trade-off than a monolithic controller.

Reinforcement Learning (RL)

Neural Network Controller for Autonomous Pile Loading Revised

no code implementations • 23 Mar 2021 • Wenyan Yang, Nataliya Strokina, Nikolay Serbenyuk, Joni Pajarinen, Reza Ghabcheloo, Juho Vihonen, Mohammad M. Aref, Joni-Kristian Kämäräinen

We have recently proposed two pile loading controllers that learn from human demonstrations: a neural network (NNet) [1] and a random forest (RF) controller [2].

Object Detection in Equirectangular Panorama

1 code implementation • 21 May 2018 • Wenyan Yang, Yanlin Qian, Francesco Cricri, Lixin Fan, Joni-Kristian Kamarainen

We introduced a high-resolution equirectangular panorama (360-degree, virtual reality) dataset for object detection and propose a multi-projection variant of YOLO detector.

Object object-detection +1